SPLAL: Similarity-based pseudo-labeling with alignment loss for semi-supervised medical image classification

TL;DR

This paper introduces SPLAL, a semi-supervised learning method for medical image classification that uses class prototypes, a weighted classifier ensemble, and alignment loss to improve pseudo-label reliability and reduce class bias.

Contribution

The novel SPLAL approach combines prototype-based pseudo-labeling with alignment loss to address class imbalance and improve semi-supervised medical image classification accuracy.

Findings

Outperforms state-of-the-art SSL methods on ISIC 2018 and BCCD datasets.

Achieves 2.24% higher accuracy and 11.40% higher F1 score on ISIC 2018.

Extensive ablation confirms the effectiveness of SPLAL components.

Abstract

Medical image classification is a challenging task due to the scarcity of labeled samples and class imbalance caused by the high variance in disease prevalence. Semi-supervised learning (SSL) methods can mitigate these challenges by leveraging both labeled and unlabeled data. However, SSL methods for medical image classification need to address two key challenges: (1) estimating reliable pseudo-labels for the images in the unlabeled dataset and (2) reducing biases caused by class imbalance. In this paper, we propose a novel SSL approach, SPLAL, that effectively addresses these challenges. SPLAL leverages class prototypes and a weighted combination of classifiers to predict reliable pseudo-labels over a subset of unlabeled images. Additionally, we introduce alignment loss to mitigate model biases toward majority classes. To evaluate the performance of our proposed approach, we conduct experiments on two publicly available medical image classification benchmark datasets: the skin lesion classification (ISIC 2018) and the blood cell classification dataset (BCCD). The experimental results empirically demonstrate that our approach outperforms several state-of-the-art SSL methods over various evaluation metrics. Specifically, our proposed approach achieves a significant improvement over the state-of-the-art approach on the ISIC 2018 dataset in both Accuracy and F1 score, with relative margins of 2.24\% and 11.40\%, respectively. Finally, we conduct extensive ablation experiments to examine the contribution of different components of our approach, validating its effectiveness.